Implementing Community-Based Diabetes Programs: The

Scripps Whittier Diabetes Institute Experience

abstrak
Diabetes affects a large and growing segment of the US population. Ethnic and racial minorities
are at disproportionate risk for diabetes, with Hispanics and non-Hispanic blacks showing a near
doubling of risk relative to non-Hispanic whites. There is an urgent need to identify low cost,
effective, and easily implementable primary and secondary prevention approaches, as well as
tertiary strategies that delay disease progression, complications, and associated deterioration in
function in patients with diabetes. The Chronic Care Model provides a well-accepted framework
for improving diabetes and chronic disease care in the community and primary care medical home.
A number of community-based diabetes programs have incorporated this model into their
infrastructure. Diabetes programs must offer accessible information and support throughout the
community and must be delivered in a format that is understood, regardless of literacy and
socioeconomic status. This article will discuss several successful, culturally competent
community-based programs and the key elements needed to implement the programs at a
community or health system level. Health systems together with local communities can integrate
the elements of community-based programs that are effective across the continuum of the care to
enhance patient-centered outcomes, enable patient acceptability and ultimately lead to improved
patient engagement and satisfaction.

Introduction
Chronic illnesses including diabetes affect a large and growing segment of the US
population. In 2009, 145 million US residents were living with one or more chronic
conditions, and 1 in 4 had more than one (1). Increases in diabetes rates reflect a myriad of influences including
higher life expectancy, population aging, more aggressive screening
practices, and marked increases in obesity and related behavioral risk factors (2). Chronic
disease treatment and management have been estimated to consume 84% of national health
care expenditures (1). Thus, there is an urgent need to identify low cost, effective, and easily
implementable primary and secondary prevention approaches, as well as tertiary strategies
that delay disease progression, complications, and associated deterioration in function in
patients with diabetes. The Chronic Care Model provides a well-accepted framework for
improving diabetes and chronic disease care in the community and primary care medical
home.(3,4) Patient self-management, delivery system design (including information systems
and decision support) and community resources are some critical components of this
approach. Substantial research shows that programs consistent with the Chronic Care Model
improve health outcomes and reduce costs among many chronic disease populations.(5)
Unfortunately, comprehensive implementation of the model components remains low in
many communities and primary care settings and practices.(6,7) Scripps Whittier Diabetes
Institute (SWDI) has developed community-based care coordination and self-management
education programs relying heavily on the role of the peer educator as a bridge between the
medical system and community. The peer educator role has facilitated improved
communications between patient, physician and health system, lowered barriers and
clarified myths related to cultural beliefs in diabetes management and ultimately improved
patient internal locus of control defined as the patients perception that the individual can
make a difference in the outcome of their own health. Over the last 15 years, SWDI has
demonstrated that this partnership with peer educators allows hospitals, communities and
physician partners to achieve better health, better healthcare, and lower cost for patients with
diabetes.
Although diabetes occurs across all ages, ethnic and racial minorities are at disproportionate
risk for diabetes, with Hispanics and non-Hispanic blacks showing a near doubling of risk
relative to non-Hispanic whites. Therefore, it is critical for community-based programs to be
delivered in a format that is understood, regardless of literacy, cultural beliefs and
socioeconomic status. Prevalence of diagnosed and undiagnosed diabetes combined is 70

80% higher in non-Hispanic blacks and Hispanics than in non-Hispanic white subjects. (8,9) More than half of
Hispanics and non-Hispanic black adults had diabetes or prediabetes in
20052006. (8) Hispanics and non-Hispanic blacks with diabetes also show poorer glycemic
control, more complications, and a higher hospitalization rate than non-Hispanic whites.
(1012) Moreover, Latinos and African Americans who are hospitalized with diabetes are at
higher risk for readmission than non-Latino whites. (13,14) Many factors, including lack of
insurance, transportation and caregiving barriers, distrust, language and cultural barriers, and
healthcare system inefficiencies may contribute to these patterns. (15,16)
If community-based diabetes programs are to make a difference in reducing the rates of
developing diabetes and improving the control of existing disease they must be able to offer
information and support throughout the community where it is accessible to all and
delivered in a format that is understood, regardless of literacy, cultural beliefs and
socioeconomic status. Several programs and key components for implementation will be
described in this article.

Community-based diabetes programs-What has been done in the past?

A number of community-based programs that employ components consistent with the
Chronic Care Model have demonstrated success and value in improving diabetes specific
outcomes. (5,1721) Patient self-management, delivery system design (including
information systems and decision support) and community resources are some critical
components in these approaches. The American Diabetes Association includes many of these components in its
annual recommendations for delivering the basic standards of care to
individuals detected with diabetes and pre-diabetes. (22). Self management education, one
of the key components of the Chronic Care Model, is associated with improved knowledge,
self-care behavior (23), and improved clinical outcomes such as lower A1C
(24,25,27,28,2932), lower self-reported weight (23), improved quality of life (26,30,33),
healthy coping (34), and lower costs (35) in patients with diabetes. Additionally, many
studies have supported the role of community health workers and peer (3642) and lay
leaders (43) in delivering diabetes self management education in conjunction with the
primary care team (44,45) A recent systematic review and meta-analysis of DSME
interventions identified an average reduction in HbA1C of .8% across 1214 mo. (46)
Consistent with the newly developing requirements and goals in the Affordable Care Act,
diabetes self management education is associated with increased use of primary and
preventive services (35,47) and lower use of acute, inpatient hospital services (35). Patients
who participate in diabetes education are more likely to follow best practice treatment
recommendations, particularly among the Medicare population, and have lower Medicare
and commercial claim costs (48,49).
Approaches using self management education can also apply to the education and support of
people at risk for diabetes, in particular in efforts to prevent the onset of diabetes. The
strategies for supporting successful behavior change and the healthy behaviors
recommended for people at risk for diabetes are similar to those for people with diabetes.
Translating evidence-based programs such as the Diabetes Prevention Program into a
community-based format can assist people at risk for diabetes in developing and maintaining
behaviors that can prevent or delay the onset of diabetes. (5052).

Project Dulce-Bringing together the components of the Chronic Care Model

into one community-based diabetes program
Project Dulce is an American Diabetes Association (ADA)-recognized care management
program developed in 1997 by the Scripps Whittier Diabetes Institute (SWDI) in
collaboration with San Diego County, federally qualified health centers, and San Diego
State University (SDSU), to improve health and access to care of underserved, ethnically
diverse people with type 2 diabetes mellitus (T2DM). Project Dulce is founded in the
Chronic Care Model (5,53), which recognizes that the majority of chronic care management takes place outside of
formal healthcare settings. Project Dulce uses a team-based approach,
with nurse care managers overseeing care, and peer educators providing culturally
appropriate, group-based diabetes self-management education (DSME), cornerstone of
effective diabetes management. (54) Peers are specially trained laypersons that serve as a

trusted bridge between patients and the healthcare system. Peer-led education and
intervention is an evidence-based method that has been shown to promote self management
and health behavior change and improve outcomes in diabetes and other conditions, and is
thought to be an effective model for reducing health disparities. (5560) Tailoring of the
Project Dulce DSME for diverse ethnic communities includes a focus on cultural beliefs, an
emphasis on the family context, and a discussion of ways to stay healthy while on a budget.
The curriculum is suitable for broad literacy levels and relies substantially on graphic
content. Content was developed with input from Hispanic, non-Hispanic Whites, nonHispanic Blacks, Vietnamese and Filipino patient focus groups and meetings with providers
to gather information on the beliefs, and barriers that affect care in these patient populations.
The curriculum acknowledges these beliefs but also creates methods to direct patients to
optimal self-management.
Studies evaluating Project Dulce have demonstrated positive effects on clinical, behavioral,
and cost outcomes, including decreased emergency room and hospital utilization.(6165) A study testing the peer
DSME component alone showed that it was effective in improving
clinical outcomes including HbA1C, blood pressure, and lipids in low income Latinos with
poorly controlled diabetes. (64) To date, Project Dulce has served more than 20,000 patients
in federally qualified health centers across San Diego County, and has been disseminated to
other health systems across the nation and in Tijuana, Mexico. An example of the
approaches used to enhance the training manual and curriculum is noted in the Figure 1.
Providing step by step instructions for the peer leaders is part of the training program
provided prior to integrating the programs into the community. Peer leaders, once identified,
undergo 40 hours of training that includes learning curriculum content, group management
skills, behavioral motivation techniques, electronic medical record navigation and workplace
skills Competencies must be met before peers can be placed in a community teaching
environment. Each section of the curriculum guides the peer educator to provide the
necessary information while allowing the flexibility to have group discussions and exchange
of personal experiences among the participants. A sample session for basic nutrition will
include teaching goals such as: the basics of nutrition; how carbohydrates and fats affect
blood sugar levels; use of the food labels; use of an action plan to set a personal goal.
Culturally tailored handouts are included at the end of the lesson plan such as traditional
foods that are high in carbohydrates/fats, types of fat, the plate method, heart healthy
eating, read the food label, dining out, weight loss tips, alcohol and diabetes, estimating
portion sizes and my goal. Finally recommended materials are suggested for each session
such as lab results to share and discuss, graphic examples (posters/models of food items),
measuring cups, spoons and standards of care cards (available from Scripps Whittier
Diabetes Institute, American Diabetes Association, American Heart Association). An
example of a patient friendly standard of care card with goals is noted in Figure 2.

Extending the community-based peer education model to other

populations
A number of excellent protocols have been established and tested that demonstrate benefit in
preventing diabetes or reducing rates of readmissions in individuals hospitalized with
diabetes. These protocols were conducted using intensive interventions that many times
were staffed with highly skilled nurses, dietitians and psychologists to deliver the programs.
In the era of cost containment in delivering health related interventions it is reasonable to
identify methods in which the interventions can be adapted and delivered effectively in
patient centric, community-based formats. The success of using peer educators within
Project Dulce has resulted in a bidirectional effect within SWDI and Scripps as a whole, to ensure that the health
system builds and develops more culturally sensitive and effective
programs whether they are based in the community, or at Scripps. This is reflected in the
subsequent development of two additional programs focused on prevention and reducing
hospital readmissions both of which are key initiatives for health systems today. SWDI has
piloted these two approaches that suggest incorporating peer-led components into the
programs are acceptable to patients and can be integrated into the existing health system to
optimize health care delivery, and potentially reduce costs.

Peer-led diabetes prevention in high risk women with a history of gestational diabetes
(GDM)
Women with a history of gestational diabetes mellitus (GDM) have a 7-fold likelihood of
future type 2 diabetes mellitus (T2DM) relative to those with normoglycemic pregnancies
(66). Latinas have the highest birth rate among all U.S. racial/ethnic groups (67) and are at
high risk for GDM (68) and T2DM (9) when compared with non-Latino whites. Lifestyle
interventions, such as the Diabetes Prevention Program (DPP), can delay or prevent T2DM onset in post-GDM and
other high-risk groups (69,70). To address the specific needs and
barriers encountered in the Latino community SWDI adapted the DPP publically available
curriculum with the Project Dulce community-based approach and conducted a single-group
pilot study to evaluate the effectiveness and acceptability of a group-format, peer-educator
administered translation of the DPP in reducing type 2 diabetes risk among Latinas with
prior GDM-Dulce Mothers. Seventy-six Latinas, 1845 years, with recent GDM underwent
an 8-week, peer-educator led, DPP-based group intervention, with adaptations for Latino
culture and recent mothers. Risk factors were assessed at baseline, month 3, and month 6.
Participants showed significant improvements in lipids, blood pressure, physical activity,
dietary fat intake, and fatalistic and culturally-driven diabetes beliefs (ps<.05). (71)
Formative evaluation suggested that the program was acceptable to participants. The
conclusions from the pilot indicted that low-cost, peer-led, culturally appropriate DPP
adaptation reduced clinical and other indicators of diabetes risk in Latinas with GDM. Based
on the preliminary results and feedback from the participants additional modifications to the
program will focus on diet and caloric balance to achieve greater weight loss.
Peer-led diabetes care coordination-Discharge to home
Hospitalizations with diabetes as a primary or secondary diagnosis increased from 2.8
million in 1988 to nearly 5.5 million in 2009. (72) Individuals with diabetes accounted for
26% of all hospital inpatient days in 2012, and these hospitalizations accounted for more
than half of the $176 billion in healthcare costs attributed to diabetes in this same year. (73)
As noted earlier, ethnic minorities with diabetes suffer more complications, a higher
hospitalization rate, higher risk for readmission and an increased mortality rate relative to
their non-Hispanic white counterparts. (10,11,13,14, 74). These failures in diabetes care
have an enormous impact on patient quality and quantity of life.
The term care transition describes the process by which a patients care shifts from one
setting to another, e.g., hospital to home. Poorly coordinated care transitions are a major
factor in hospital readmissions, complications, and worsening health. (75,76) Failures of
care coordination including poor care transitions accounted for up to $45 billion in wasted
healthcare spending in 2011. (77) Further, three quarters of short-term hospital readmissions
are potentially preventable. (78) Care transitions have thus been identified as a critical target
for improving healthcare quality. (76) Indeed, the Affordable Care Act will fund
implementation of evidence-based care transitions services for Medicare patients at high risk
for readmissions (Section 3026 of HR 3590). In addition, under the Affordable Care Act,
Medicare recently began financially penalizing hospitals that have higher-than-expected 30- day readmissions rates
for some conditions. A recent systematic review of 24 studies testing
structured discharge programs showed positive effects on readmissions and length of stay,
with some studies suggesting improved patient satisfaction with care. (79) Research
concerning the effects of such programs on other patient relevant health outcomes (e.g.,
functional status, well-being, perceived health, quality of life) and on clinical control was
noted to be very limited, highlighting an important gap in evidence. Indeed, the review
highlighted the need for additional, rigorous research examining such programs. Further,
their reliance on nurses, pharmacists, or other professionals, means that the programs
reviewed entail a resource intensive commitment in new staff or set aside time for existing
staff. Moreover, such programs may not adequately address the specific needs of lowincome,
ethnic minority patients, or patients with diabetes, who are at especially high risk
for readmissions. A recent report that gathered perspectives of leaders from state and local
hospitals and improvement teams noted that many hospitals have eschewed discharge
planning programs because they do not view the benefits as outweighing the costs, and they

feel that there is inadequate evidence regarding cost-effective interventions to guide improvement. (80) Additional
research is needed to fill these evidence gaps and better guide
provision of care and care improvement for the highest risk populations.
SWDI recently developed and piloted a novel care transitions program for high-risk patients
hospitalized with diabetes. The Dulce Transitions program assigns specially trained peer
educators (peer transitions coaches) to follow patients discharged from the hospital,
connect them with outpatient health and other community resources, and activate them to
effectively manage their disease within their own environment and cultural context by
developing necessary knowledge skills, self-efficacy, and resources, and overcoming
barriers (Figure 3). The pilot program reached 200 patients between October 2010 and
February 2012. An internal evaluation revealed a reduced 30-day readmissions rate for these
patients of 11% versus 16% in similar patients who were not included in the program during
the same period. This 31% reduction in readmissions is comparable to that achieved in nurse
led programs, such as the successful care transitions program developed by Coleman and
colleagues (81) and the Re-Engineered Discharge program. (82) Clinical and other patient
relevant outcomes were not tracked in this pilot. Patients and peer providers were surveyed
to obtain their impressions of the programs strengths and areas for improvement. One
patient from Scripps Mercy Hospital Chula Vista reported being hospitalized up to 2 times
per month prior to participating in the Dulce Transitions pilot. The peer-led diabetes selfmanagement
program resulted in knowledge gained by the patient and a positive outlook
towards her health and future.
In both examples further studies will be required to formally test the effectiveness of
communitybased peer led models in improving clinical and patient centered outcomes
(e.g., quality of life) compared to a nurse-led discharge support intervention or other models
of prevention programs. Optimally this should be developed in a partnership model that
integrates patients and other stakeholders from inception to dissemination.

Key strategies for implementing community-based diabetes programs

Consistent with the Chronic Care Model, key components must be in place in order to successfully implement a
community-based diabetes program into specific communities.
The key elements of the underlying infrastructure might be considered a checklist that
must be completed before deployment can occur.
Key elements include the following:
Data collection and tracking system-The system must collect and report on the
outcomes that the community or health system would like to manage to achieve
improvement. The most common elements include glucose (HbA1c), lipid (LDL)
and blood pressure control, but can also include vaccination requirements
(influenza, pneumonia, hepatitis B), smoking status, aspirin use, depression and
diabetes distress scores, and other self care management scores. Other long term
outcomes that may need to be tracked and reported includes annual costs related to
specialist visits, emergency and urgent care visits and hospitalizations. The data
tracking and reporting system can be a simple registry specific for diabetes
outcomes or a more sophisticated data analytics program that is tied to the
electronic medical record or to a health information exchange that captures the
required elements from all sources of patient care and interaction.
Staffing-A community-based program must have clinical and operational leadership
to develop and sustain an ongoing program. A physician champion who is either an
endocrinologist or primary care physician knowledgeable in the diabetes standards
and literature can take on the role of medical leadership. They will provide guidance and support for the clinical
decision support tools and be available as
back up for complicated clinical situations. One or more certified diabetes
educators will provide oversight for maintaining the integrity of the clinical
interventions and the self management education programs. A program manager
will have oversight responsibilities for staffing, program location, integration with
existing community and health system services, distribution of tracking reports and
communication between program staff, health system, community and patients on
the program progress. Underlying support services may be required from finance,

human resources and information technology to maintain smooth operations of the

program. The program leadership will need to decide what type of staff mix will be
needed to deliver the program content such as nurses, dietitians, health coaches or
peer educators.
Training-It is important to ensure that all who that will be participating in the
program delivery are appropriately trained. Physicians, health system operations
staff and community leaders involved in the program should be knowledgeable in
the basic concepts of the Chronic Care Model and team-based care. Staff who are
directly responsible for the development and implementation of the program should
be trained in the basic operations requirements, job descriptions and competencies
to conduct each role. Staff who will be delivering the content must have training in
the content delivery and provided with ancillary materials for patient education that
is appropriate for the language, literacy and culture of the community where the
program will operate.
Health systems integration-A community-based program must ensure that it can
share summary reports, labs and other data with the local health systems and
physician offices that it will be serving. A shared electronic medical system can
facilitate this process but is not always easily available therefore other
communication plans must be established.
Local resource identification-Identifying additional community resources that will
assist patients in achieving their clinical and behavioral goals are critical since
these systems may be more easily available to patients in their neighborhoods.
Community resources can include local community health centers, county health resources, senior centers and
YMCAs.
Ongoing communication among all parties-Regular meetings should be planned to
resolve any operational and clinical issues, to discuss outcomes and to ensure that
there is bidirectional dialogue between all the constituents of the program including
patients and community leaders for recommendations and ongoing improvements.
At SWDI we have found that integrating peer educators into the glucose management team
as part of community-based diabetes programs has been a key to successfully breaking down
cultural and socio-economic barriers to improved care. Appropriate, standardized training is
required to ensure the highest caliber of program delivery. In return for the significant effort
made by the peer in training a stipend is offered after completing their basic training.
Competencies must be met before peers can be placed in a community teaching
environment. Once the peer educators complete the training and competencies they are
considered eligible for employment and are compensated to conduct the Project Dulce
DSME programs. Type of employment can vary and may be contracted or full-time. Peer
educators can be hired by the health system, community health centers or other community
based organizations to conduct programs depending on the needs of the individual
community. At SWDI peer educators are integrated into the health system glucose
management teams so that on-going support is provided with licensed professionals. Conclusions
The population that requires knowledge and access to diabetes programs is growing
worldwide. We must be able to offer information and support throughout the community
where it is accessible to all and delivered in a format that is understood, regardless of
literacy and socioeconomic status if we expect to make a difference in reducing the rates of
developing diabetes and improving the control of existing disease. The Chronic Care Model
offers a framework for creating and implementing community-based programs. The
American Diabetes Association Standards of Care provides additional recommendations and
guidelines to support program development. (22) Strategies such as explicit goal setting with
patients; identifying and addressing language, numeracy, or cultural barriers to care;
integrating evidence-based guidelines and clinical information tools into the process of care;
and incorporating care management teams including nurses, pharmacists, and other
providers have each been shown to optimize provider and team behavior and thereby
catalyze reduction in A1C, blood pressure, and LDL. At SWDI, our community programs
have been anchored by peer educators who have been trained to deliver content and serve a
bidirectional role in strengthening the nterivention as well as delivering it. Finally, scaling

and replicating local programs that have been tested and evaluated can offer options for
integrating community-based programs across the continuum of integrated delivery
networks to enhance patient-centered outcomes and enable patient acceptability ultimately
leading to improved patient engagement and satisfaction. (8392)